Desktop Metal Manufacturing Services: A Quality Inspector's 7-Step Approval Checklist

If you're evaluating desktop metal manufacturing products and services for your shop, you're probably drowning in vendor content. Most of it is either a pitch or a review written by someone who never touched the parts. I have.

I'm a quality compliance manager at a custom manufacturing company. I review every part that reaches customers—roughly 250 unique items a year. In Q1 2024, I rejected 14% of first deliveries from new suppliers due to tolerance non-conformance. Not because I enjoy sending parts back. Because it's a lot cheaper than explaining a field failure.

Quick note before we start. You know how people search "fraxel versus co2 laser" when they're actually trying to decide between two very different skin treatment tools? The same mix-ups happen in metal manufacturing. Desktop metal printing, a desktop metal lathe, CNC milling machine types, an Italian press brake—they're all different tools for different jobs. Mixing them up while evaluating a supplier will lead you to approve the wrong process.

This checklist has seven steps. Do them in order, and you'll catch most problems before they cost you a redo.

Step 1: Match the Process to the Geometry—Not the Brochure

"Desktop metal" covers a lot of ground: binder jetting, laser powder bed fusion, directed energy deposition. A production binder jet system like the Desktop Metal P-50 handles high-throughput parts differently than a laser powder bed machine. The trade-offs matter.

Your first question to a supplier shouldn't be "can you make it?" It should be "what process do you recommend for this geometry, and why?"

If your part has internal channels, thin walls, or density requirements, binder jetting has a critical follow-up: parts get sintered after printing, which means controlled shrinkage. A good supplier explains that up front—not after you discover it in your inspection lab.

We learned this the hard way. I approved a prototype batch with complex internal cooling channels without asking which machine it was slated for. The vendor had shown us gorgeous sample parts. The geometry didn't suit their printer. We got dimensions all over the place. Actually, the variation was within their claimed range—but well outside ours. Looking back, I should have requested the build report in the same email as the RFQ. At the time, it felt like an unnecessary show of distrust. It wasn't.

Step 2: Demand Material Traceability Before You Approve Anything

This sounds obvious, but it's the most common gap I find in supplier audits.

Ask for material certifications, lot numbers, heat numbers—and I take "confirming the chemical composition matches the spec" as a given.

In 2023, we received a batch of 800 units where the vendor's certificate showed the right alloy, but their internal lot tracing had gaps. We ran our own composition test and found inclusions that should have been caught at incoming inspection. The vendor claimed it was "within industry standard." We rejected the batch. They redid it at their cost. Now every contract includes material traceability requirements, and we audit their incoming material logs.

If a supplier can't trace it, don't ship it.

Step 3: Perform Your Own First-Article Inspection

The supplier's inspection report is a starting point. Not a conclusion.

When first articles arrive, we measure everything on the drawing—every GD&T callout—using our own CMM and surface tester. Why our own? Because I've seen a supplier pass their own parts while their height gauges had overdue calibration stickers.

That was 2024. I still kick myself for not checking their calibration paperwork during the initial audit. If I'd caught it then, we'd have saved a three-week correction cycle and a late launch. Now, calibration status on every new supplier's measurement equipment gets checked. Every time.

Step 4: Plan Post-Processing Early—Not After Parts Arrive

This is the step most people ignore.

Additive processes rarely finish a part. Binder jet parts are porous before sintering. Powder bed parts need support removal and often heat treatment. And functional surfaces almost always need machining.

Get these answered in the quoting stage:

• Where are the support contact points, and how will supports be removed?
• How much machining stock is left on critical surfaces?
• Who owns deburring and surface prep?

We once ordered a run of manifolds where the sealing face didn't match the machining datum. The vendor chose a different reference, and we lost about 0.015 inches on the seal. That was a $22,000 redo and a six-week slip on launch. Post-processing isn't a detail—it's where quality gets decided.

Step 5: Know Which CNC Milling Machine Type the Job Actually Needs

Additive doesn't do everything well. Dowel pin holes, sealing surfaces, threads—those need subtractive machining.

CNC milling machine types matter more than people expect:

• 3-axis: the workhorse for prismatic features and flat faces.
• 4-axis: adds rotation, so you can index around a part in one setup.
• 5-axis: handles complex contours and undercuts in fewer setups.

If your part has round features—O-ring grooves, bearing seats, axisymmetric profiles—a desktop metal lathe is often the tighter, faster choice than milling those features. Turning gives roundness that milling can't easily match. Ask your supplier: "Are you milling this groove or turning it? What roundness are you holding?" If they can't answer without hesitation, that's a red flag.

Step 6: Tour the Whole Facility, Not Just the New Machine

When I audit a supplier, I want to see the entire floor. In person is best; a live video walkthrough works if you're remote.

One machine doesn't make a manufacturing operation. The shops I trust have a complementary mix: CNC mills, surface grinders, a lathe or two—even a desktop metal lathe for small precision work—plus sheet metal capacity like an Italian press brake. That mix tells me they understand the full sequence: forming, machining, inspecting.

An Italian press brake, mind you, has nothing to do with additive printing. But it tells me the shop understands bending limits, bend allowances, and how metal behaves under stress. That knowledge carries over to how they handle your parts.

Step 7: Start With a Small Batch

I'll be honest with you: this advice gives me mixed feelings, because it sounds like I don't trust anyone. The opposite is true. A small order is a test drive.

In 2024, we qualified two new suppliers for a product redesign. We started with $300–$500 prototype orders—parts with the same tolerances we'd demand in production. The suppliers who treated those small orders seriously, who returned clean first-article reports and answered the phone when we had questions, are the ones we now send $20,000+ production orders to.

One vendor turned down our $300 order entirely. "Minimum order quantity is $5,000." Their right. But it told us everything we needed to know about how they treat small customers.

Today's small customer is tomorrow's larger account. The vendors who get that tend to get my business.

Common Mistakes That Cost You Money

• Approving based on marketing samples. Printed smiley faces aren't sealing faces. Ask for parts with features similar to yours.
• Skipping material certs. If they can't trace it, don't ship it.
• Treating post-processing as an afterthought. It'll come back to haunt you as a change order.
• Forcing additive to do everything. Additive and subtractive are complementary. A tolerance-critical dowel hole is a machining job, period.
• Ignoring calibration status. Most inspection surprises I've seen trace back to a gauge that wasn't calibrated.

Bottom line: desktop metal manufacturing products and services can deliver serious value—when you evaluate them with a checklist, not a brochure. Match the process to the geometry. Trace your material. Inspect first articles with your own instruments. Plan post-processing early. Know your CNC milling machine types. Tour the whole facility. And start with a small order.

Do that, and the risk drops. Skip it, and you'll learn the hard way—like I did, more than once.